aws-architecture-diagram
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's behavior is consistent with its stated purpose of generating and styling AWS architecture diagrams. It utilizes platform tools (Bash, Write, Read) to analyze infrastructure code and save outputs. Preview functionality uses the official diagrams.net web application to render the generated XML, which is standard for this toolset.
- [INDIRECT_PROMPT_INJECTION]: The skill presents an ingestion surface for untrusted data during its codebase analysis phase. 1. Ingestion points: Infrastructure files (CloudFormation, CDK, Terraform) are scanned in the codebase analysis mode (SKILL.md). 2. Boundary markers: Absent; the skill does not explicitly use delimiters for scanned content. 3. Capability inventory: Uses the Bash tool to execute local Python fixers and preview utilities, and the Write tool to save diagram files (SKILL.md). 4. Sanitization: The post-processing pipeline uses the defusedxml library to prevent XML External Entity (XXE) and other parsing vulnerabilities (references/post-processing.md). This risk is low and inherent to the skill's primary function of infrastructure analysis.
Audit Metadata